Rnjiamycim
`
`63
`
`Suppression of Allograft Rejection by
`Combined Treatment With CsA, FK506, and
`MPA
`Until recently, formal pharmacological principles
`had rarely bccn applied to clctrrminc:- whether sinrnl(cid:173)
`tancousl)' administered immunosuppressants pro(cid:173)
`duce a net state of immunosuppression in graft
`recipients that is antagonistic, additive, or ~-ynergistic
`compared with immunosuppression caused b)' the
`individual administration of each agent. For many
`years, Berenbaum has decried the expe1imental
`design and data analysis from studies in transplanta(cid:173)
`tion, among other fields, that have led to incorrect
`conclusions concerning immunosuppressivc drug(cid:173)
`drug interactions.w1 Several years ago, when wf."'.
`converted the standard mouse car-heart transplant
`technique to a quanta! bioassay, we tried to redress
`the inadequacies of pre\~Ous studies of combination
`immunosuppressive therapy. 17
`When we first began investigating the immunosup(cid:173)
`pressivc activity of RPM in vivo, we assumed that the
`structural similarity between RPM and FK506 pre(cid:173)
`dicted that both drugs affected the immune system
`ve11' similarly because at that time there was no in
`vivo or in vitro 'data to the contrary. 111is nssumption,
`combined with our previous finding that treatment
`with FK.506 docs not antagonize CsA immunosup(cid:173)
`pression in vivo,11 led us to treat rat heart allograft
`recipients with minimally effective doses of RPM plus
`CsA.3 This study was not sufficiently rigorous to
`enable us to conclude from our data that these two
`dru~ interact to produce immunosuppression that is
`synergistic. However, we were able to show that this
`combination is not antagonistic and that the immu(cid:173)
`nosupprcssion caused by this combined therapy is at
`lea~t additive. A recent and extensive study! involv(cid:173)
`ing sixteen separatr treatment groups that ex(cid:173)
`panded prc,~ous work!11 clearly showed, that com(cid:173)
`bined treatment with RPM plus CsA produces
`synergistic suppression of rat heart allograft rejec(cid:173)
`tion. A smaller subset of this study showed that
`combined treatment with RPM and CsA is also
`beneficial for the prolongation of rat kidney al(cid:173)
`lografts.
`Using the mouse car-heart bioassa)', we showed
`that multiple combinations of RPM plus CsA or
`RPM plus FK506 cause prolongation of graft sun~val
`that is synergistic as defined by isobologram analysis''
`when the treatment doses of each drug are less than
`their ED,.,s. In addition, probit analysis has been used
`to show that combined treatment or mouse skin
`
`allograft recipients with RPM plus CsA produces
`immunosupprcssion that is syncrgistic.3; Furthcr(cid:173)
`morr, high doses of both RPM and FK506 used in
`combination did not indicate in any way that either
`dnig antagonizes the immunosupprcssive effects of
`the other. The data showing that treatment with
`RPM plus FK506 produces synergistic immunosup(cid:173)
`prcssion at many close levels contradict studies in
`vitro that showed these two drugs antagonize each
`other's effects on immune cells (discussed in section
`ht:aded Effects of RPM on Immune Cells in Vitro).
`Lately, we have extended our interest in combina(cid:173)
`tion immunosuppressive drug therapy to .the com(cid:173)
`bined use of three dn1gs that have dilferent mecha(cid:173)
`nisms of immunosuppressivc action (Fig 3) and
`nonovcrlapping toxicity.9 For example, low-dose CsA,
`RPM, or l\tlPA (administered as its prodrug, RS-
`61443) monotherapy ineffectively prolongs rat heart
`allograft SUr\~val (Table 11 ). When these same doses
`are used, but all three drugs arc administered to(cid:173)
`gether, suppression of graft rejection is not only
`more effective than when each drug is used st:pa(cid:173)
`rately, but it is also more effective than when RPM
`plus .MPA or CsA plus MPA is used. Additional
`follow- up is required to determine whether triple
`drug therapy is superior to treatment with RPM plus
`CsA.
`RPM monotherapy of cynomolgus mopkcy heart
`allograft recipients prolongs graft sU1vival (discussed
`previously). However, unlike the use of RPM in
`rodent graft recipients, monkeys seem more resis(cid:173)
`tant to the immunosuppressive effects ofRPl\·i and
`more sensitive to its toxicity. Because RPM and CsA
`produce immunosupprcssion in rodent graft rec-ipi(cid:173)
`cnts that is ~·ypergistic, and because it is possible that
`the to_xic effects qf each drug are different, we
`treated monkey heart allograft recipients (Morris
`RE, Wangj, Shorthouse R, et al: unpublished obser(cid:173)
`vation, 1991) \\~th low doses of both drugs for the first
`JOO clays posttransplant (Table 10), This treatment
`regimen suppressed r~jection much more dfectively
`than treatment with either CsA or RPM monother(cid:173)
`apy. Only two of the five monkeys treated with
`combination therapy rejected their heart allografts
`during the treatment period; all animals remained
`clinically well. Pharmacokinetic analyses of CsA blood
`levels showed that the 2 mg/kg dose ofCsA produc-es
`CsA levels that arc subtherapcutic: (all <60 ng/mL).
`Concomitant treatment with RPM and CsA docs not
`elevate CsA blood levels compared with levels at(cid:173)
`tained with CsA treatment alone. Thus, the im-
`
`
`
`West-Ward Exhibit 1055
`Morris 1992
`Page 0026
`
`

`

`64
`
`R1111d11// £/Iii .1/oniJ
`
`Table 11. Ellt-r1 of' Combination Therapy "ith LU>i\1, CsA and ;\l)'Cnphenolk Add (as ils 111orphoincL11yles1cr, RS-6 1-1-11
`un tht' Su1vival of'B1 uwn-:\01way Rat Heternlllpk Abdomi11nl H ~art Grafts Tmnspl::1111cd i11to Le\\ is Rl·cipienls
`
`)
`
`'Drug(s)
`
`GsA
`Cv\
`i\IPA
`IU';\l
`lU':0-1
`+ i\JPA
`CsA
`+l\JPA
`IU'i\1
`+CsA
`CsA
`+ l\lPA
`+RP1'1
`
`DoJf!
`(111g/ l.gJ
`
`Rnu/f!
`
`For11111/ntin11
`
`Sdmlule
`(dt!)I)
`
`.l!rdia11 Grqfi
`(da.J"I)
`
`0.5
`
`0.75
`
`10
`1.5
`1.5
`10
`0.75
`10
`1.5
`0.5
`0.5
`10
`1.5
`
`£P
`
`£P
`PO
`PO
`PO
`PO
`
`II'
`PO
`PO
`IP
`
`IP
`PO
`PO
`
`Solution in C:mnophor EL/ethaool
`Solution in Cremophor EL/ethanol
`
`su~pt'l1>iun in ca1 boxymethrl cellulose
`Smpension in mrboxymcthyl cdlulosc
`Suspension in carbnxpnethrl cellulose
`Suspc11sion in carboxy111e1hyl cellulose
`Snluurm in Crcmuphor l::Uethanol
`Suspension in carboxyme1hyl cellulose
`Suspension in carboxymethyl cellulose
`Solution in Cmnophnr EL/ethanol
`
`Solution in C:remophor EL/ethanol
`Suspension in carboxymethyl cl'lluln.~I'
`Suspt·n~1on in c;u bo.'\)11lcth) I cellulose
`
`I to50
`1 co50
`1 to50
`
`I to.'iO
`
`1 to50
`I 11150
`1 rn51l
`I lu50
`
`I 1050
`I to50
`1 to50
`I to50
`I to50
`
`9
`15
`10
`11
`
`28
`
`67
`
`109
`
`170
`
`proved immunosupprrssive efficacy caused by combi(cid:173)
`nation therapy cannot be explained by high CsA
`blood levels (data not shown). The coadmi11is1ration
`ofCsA could elevate RP/\{ levels, but "~thout a blood
`level assay for RPM, this possibility cannot be exam(cid:173)
`ined. In the raL, coadministration of RPM and CsA
`docs not clcvntc C3A levels-"'
`FK506 is known lo suppress hepatic cy1oduo111c
`NSO and Lhc acth~ties of ethylmorphine 1\-cl<'mcl h(cid:173)
`ylasc and cytochrome c reductase in rats,"'1 and this
`may partly accoun1 for the increased half-life of CsA
`in patients treated \\ith FK506."1 Because we did not
`line! that RJ':-.1J increases tht haJf. Jjfc of CsA in
`monkeys;• the effects of fK.506 and RPJ\I on the
`m etabolism of CsA may differ. The i111 rraC'tinns
`between nonim 111unosuppressive macrolidc antibiot(cid:173)
`ics and CsA have been defined and mar prO\·icle
`additional clues to the interaction between RPM and
`CsA.1111
`·n1csc initial studies of Lhl' lack of effect of
`treatment with RP~L on CsA blood levels sugg<'Sl
`that the combined USC of RPM plus Cst\ ma} anorcl
`the benefits of increased immunosupprcssi\'e efficacy
`\\~thout the penalty of decreased safety. In view or
`the similar mrrhanisms ol' immunosupprcssi\'C ac(cid:173)
`tion or CsA a nd FK5UG, the supt'l'ior cllicac->· and
`potency of FK.506, and lhc synergistic i111 mu11os11p(cid:173)
`prcssion caused by the adminis tration of RPM phis
`FKSOG in mouse car-heart recipients," the combim·d
`usr oflU'M and FK506 may a lso be useful in monkry
`and human graft recipients. H owc,·cr, the incrcilsccl
`nephrotoxicil~ and cliabctogenic dli-ct or eombinccl
`
`high doses of RPM plus CsA in the rat'~ alerts us to
`the possible synergistic toxicit}• that can be caused by
`specific drug combim1lions.
`
`Interactions Between RPM and
`Nonimmunosuppressive Drugs
`RPM is likely to be used in patients receiving compli(cid:173)
`cated treatment involving a \\ide variety of nonimmu(cid:173)
`nosuppressivc drugs. Jn addition to the interactions
`of RPM \\ith other immmunosupprcssants (previ(cid:173)
`ously discussed), coadministration of nonimmunosup(cid:173)
`pressivc drugs may also subslantially inRuence our
`gual or optimizing the dose, route, and schedule or
`aclminis1nuinn uf RPM. Because RJ>l\·I shares some
`physirochemkal characteristics with CsA, and is
`structurally similar to FKSOli, thr extensive experi(cid:173)
`ence of Cs.A. drug interactions"'' and the increasing
`understanding of the pharmacology of FK.506''"""
`may prO\idc lessons that will nor ha,·c to be com(cid:173)
`plctcl)• relearned \\ith RPi\l. If, like CsA and FK506,
`RP~l blood ll'vcls and its pharmacological effects
`\'ary widely among patients, and if' the therapeutic
`index or RJ';\[ is low, the effects of simultaneously
`administered drugs will profoundly affcl·L t he clinical
`use ol'RPi\'L
`As descri bed in a rcccnl review on C:sA drug
`intcractions,"'1 1his potc:nli::tll)' complex problem can
`be simplified, at least cone'' Pl uallr, b)' <rnalyzing how
`drug interactions affect Lil<' absorption, distribu1io11,
`metabolism, and elimination (pharmacokinrtics) of
`drugs and their the biolocial/toxicological effects
`
`
`
`West-Ward Exhibit 1055
`Morris 1992
`Page 0027
`
`

`

`65
`
`(pharmacodynamics). For example, coadministra(cid:173)
`tion of drugs that affect bile Aow and gaMrointrstinal
`dysfunction shnuld alfcct the absorption of orally
`administered RPM. The physicochcmical similaritii:-s
`among RPt\11, CsA, and FK506 also suggrst that thr
`distribution ol'RP.M in tissues and cells will be similar
`to CsA and FK506. Although the metabolic pathways
`for RPM arc unknown, the structural similarity of
`RPM with FK506 woulcl suggest that RPrvl is pri(cid:173)
`marily metabolized in the liver. Once the metabolic
`pathways in the liver for RPM have been drfined, we
`will be able to predict how.other drugs might alter its
`metabolism. For example, there is extensive informa(cid:173)
`tion on how inhibitors and inducers of hepatic en(cid:173)
`zymes associated with the cytochrome P450 system
`affect the metabolism ofCsA.'" 1
`CsA and FK'l06 arc nephrotoxic at immunosup(cid:173)
`prcssivc doses, and RPM is not completely without
`the potential to cause nepluotoxicity.''' Therefore,
`drugs that are known to exacerbate CsA- and FK506-
`induced nephrotoxicit/" should also be evaluated
`for their ability to unmask nephrotoxic effects of
`RPM.
`
`Effects of RPM on Cells and Tissues
`of the Immune System ill Vivo
`For RfM to suppress allograft rejection or :iutoim(cid:173)
`m'uni:- diseases effectively, it must alter the normal
`!Unctions of tl1e immune system. Severn! studies
`using the most advanced techniques in cellular immu(cid:173)
`nology and molecular biology have been intelligently
`exploited to try to understand how RPM affects
`immune cells under rigidly defined conditions in
`vitro. On one hand, these highly controlled experi(cid:173)
`mental systems prm~de relatively dean answers to
`significant questions about the effects of a drug on
`very specific immune functions; on the other hand,
`important drug effects that fall outside the necessar(cid:173)
`ily narrow focus of these in,·cstigations can be com(cid:173)
`pletely overlooked. Even when in vitro studies are
`focused appropriately, the answers that these experi(cid:173)
`ments pro-vide may not always be relevant to mecha(cid:173)
`nisms or immunosuppressant drug action in vivo?1
`Changes in drug blood level, drug binding to plasma
`proteins, conversion of the parent drug into acti\'c
`and ibactive metabolites, and the complex micn.ienvi(cid:173)
`ronmcnt oftluctuatingeytokine levels that character(cid:173)
`izes the response of the immune system to antigen in
`vi\'O cannot be duplicated in vitro. Therefore, beforr
`examining the effects of RPM on approximations of
`the immune system in vitro, wr will rc\~CW\\'hat little
`
`is known about the effects of RPM on cqmponents of
`the immune systen1 in vivo.
`
`Suppression of the Host-Versus-Graft and
`Graft.Versus-Host Responses
`
`The host-versus-graft (HvG) popliteal lymph node
`(PL!\) assay approximates the mixed lymphocyte
`reaction in vitro. We found that treatment with
`RPM, CsA, or FK.506 suppresses I he increasr in PLN
`weight caused hy the irtjection of irradiated BALB/c
`spleen cells into thr hind feet ol'C3H mice.& Ongoing
`studies using Aow cytometric and in situ hybridiza(cid:173)
`tion analyses are designed to determine whether
`these drugs suppress the response of the PLN to
`alloantigcn by inhibiting cytokine gene transcription,
`cytokine synthesis, cell proliferation, migration of
`cells into the node, or by promoting the exit of cells
`from the node (Morris RE, Shorthouse R, Zheng B,
`ct al: unpublished observations, 1991). Despite the
`superior potency of RPM in the mouse ear:heart
`bioassay, RPM inhibits thr HvG response less po(cid:173)
`tently than FK506. This finding suggests that the
`exceptional efficacy of RPM for prolongation of
`rnurine heart graJ't survival may be more complex
`than can be accounted for b)' the PLl\I assay.
`Recently, the in vitro function of cells in PLN
`draining the foot pads of mice t hat had been injected
`with allogencic cells has been examined.'f\i In con(cid:173)
`trast to mice treated \\~th CsA or FK506, the PLN
`cells from mice treated with RPM incorporate less
`thymidine spontaneously or when stimulated with
`intrrleukin-2 (lL-2). Cells from RPM-treated mice
`aL~o arc less capable of generating cytotoxic T-ccll
`acti\ity or natural killer cell activity than cells from
`PLN from mice treated \\~th either CsA or FK'i06.
`The PLN assay can also be used to approximate
`the graft-versus-host response (GvH). We found that
`RPM is able to suppress this response in mice.';These
`results suggest that RPM may have the potential to
`control this disease in recipients of allogeneic bone
`marrow transpla11ts. Further speculation Jed us to
`propose that RP!'vl be used to facilitate engraftmcnt
`of bone marrow derived from organ donors and for
`the creation or a chimeric state for induction or
`donor-specific unresponsiveness in human graft recip(cid:173)
`ients.''
`
`Effects on Numbers and Immune Function of
`Peripheral Blood T and B Cells
`As part of a subchron.ic toxicity sludy of RPM in mice
`(unpublished), total WBC counts were monitored in
`mice thal had been treated [p daily for 14 days with a
`
`-
`
`
`
`West-Ward Exhibit 1055
`Morris 1992
`Page 0028
`
`

`

`66
`
`Uandllll Ellis tlloniJ
`
`very high dose (2+ mg/ kg) of RP~l in suspension.
`·n1e WBC counts arc normal on dar I+, but during
`the rec:ovcry period it was found that this dose or
`RPM depresses the \o\'BC counts. llowc1·c1-. artc r 2
`weeks of daily 11' treatment of rat heart rcripirnts
`with 6 mg/kg or RP:\1 (a dosr that produces > 200
`day gr.ift survival for all grafts), the WBC and the
`rotal lymphOC}te counts arc normal. Thus, indefinite
`prolongation of graft survival in rats occurs after
`brief RPl-1 treatmcnt \\ithout depiction of circul<ll(cid:173)
`ing lymphocytes. Although RPM can moderately
`suppress the WBC count in mice, this dTrtt is by nn
`means sufficient lo explain how RP!Vl induces indefi(cid:173)
`nite prolongation of graft su1vi1<1I. The rejection of
`third-party grafts transplnnted into recipients bear(cid:173)
`ing viable primary grafts (discussed in section headed
`Effects of RP~J on Graft and Tissue Rejection) not
`only indicatt's that there arc sufficient numbers of
`cir.culating lymphocytes lo mediate rejection, but
`also indicates that these cells arcsclcctivcl)• immuno(cid:173)
`competent.
`Within 2 weeks of treating monkeys with RPM
`monotherapy or with combination therap)' of RPM
`plus CsA, the absolute lymphocyte count is lowered
`(Morris RE, W:u1g J, Zheng B, et al: unpublished
`obse1vations, 1991). Flowc)'1:ometricanalrsis ofprep(cid:173)
`nrntions of monkey whole blood showed (Mo,.ris RE,
`Wang j , Zheng B, cl al: unpublished obsc1valions,
`1991) that the Lolal numbe rs of both T and B cells
`are lower than pretreatment va.lues. In untreated
`monkeys, the ratio of the number of CD8+:CD4+
`rclls is greater than unity. RPM hcatment causes
`this ratio to become inverted, because there is a
`disproportionate reduction in the number of COil+
`cells compared with CD4+ cells. It docs not seem
`that t he alterations in cell number alone can account
`for the suppression of graft rt'jcction caused b)' RPM.
`More likely, RPM produces immunosupprcssion b)'
`functional!)• inacti\'aling immune cells. Fur example,
`when WC quantitatcd the rcsponsc Of peripheral
`blood mononuclear cells in monkeys treated with
`RPlVf plus CsA to differen t concentrations or con(cid:173)
`canavalen A (ConA) in vitro, we found that mi to(cid:173)
`genic responses arc suppressed al low concentrations
`of ConA but return Lo levels similar to prctreatmenL
`\'alucs a~ the concentration of ConA in culture is
`increased (Morris RE, Wang J, Zheng 13, ct al:
`unpublished obsc"".itions, 1991). Changes in the
`numbers and function of circulating peripheral blood
`B cells, T cells, and T-ccll subsets caused by RJ'l\lf in
`the monkey ma)' also occur in humans Lreated with
`RPM. iJ so, these parameters may provide a more
`
`sensit.ht' index of the effects of RPM on the immune
`system than assessment of graft rejection.
`
`Effects on th e Morphology and Function of
`Central Lymphoid Tissue
`
`Thr only hint uf the rational<' for the firs l investiga(cid:173)
`tion of the immunosupprcssiw rffects of RPM by
`l\fartel ct a1 ~1 was a brief sentence in the Discussion
`section of their article which read, " .. .long-cerm
`toxicity studies in dogs (Hemm RD, Au thicr L:
`pc1 son al communication) have clcmonstratcd that
`rapamyri1i caused hypoplasi;i of lymphatic: tissues
`(l}-mph nodes, splren, and chymus)." This effect of
`RPM has now been confirmed in other species. For
`example, as part of an initial subchronic toxicolog)'
`scudy we treated mice IP daily \\'ilh a dose of RPM
`(24 mg/kg) that far exceeds doses (6 mg/kg) re(cid:173)
`quired to prolong car-heart grafts indefinitely. ~ci.;­
`ropsies (Morris RE: unpublishc;.d obsc1vations, 1989)
`of mice on clay 14 showed the thymus to be dramali·
`caUy involuted, but the lymph nodes and spleen
`seemed normal in size and weight. 1\.Iicroscopic
`analysis of the lymph nodes and spleen did not show
`any abnormalities, but the normal distinction of the
`th)1nic cortex from the medulla was often absenl
`and Lhymic lymphoid depiction was profound. When
`other animals from che same rrcaiment group were
`nccropsicd on clay 25 (2 weeks after the Inst RP~l
`close), thymic involution persisll:d and lymphoid cells
`were decreased in the mcdulla1y cords of lymph
`nodes.
`A more thorough study"" was conducted in mice
`that were treated IP daily with 6 mg or. 75 mg/kg or
`RPM for a maximum of 13 days. These mice and
`aged-matched control mice were necropsied 011 days
`7, If, 21, 4·2, and 102, and thei r thym us and spleen
`weights recorded. Tissues from Lhe thymus and
`spleen \\ere stained with monoclonal antibodies and
`anal}7.t:d by immunohistochemistry and Aowcytomc(cid:173)
`ll) .. Fi nail)•, spleen cells were cultured and stimulated
`\vi th inc1 casing concentrations of either the T-cell
`mitogcn ConA, or the B-ccll 111itogcn Sa/111011ella
`01J11i11111riw11 (STl\>l).
`We found that RPM treatment docs not decrease
`the weight of the spleen. In contrasl, RPi\l treatment
`has complex cffrcts on the weight of the thymus.'"
`111e 6 mg/ kg dose of RPM cuuscs the thmus weight
`to be reduced b)' 80% after I week of treatment. The
`thymus "eight increases, but is still abnormally low
`bydn)'42; by day 102, the weight n·bounds w normal.
`T he lower dose or RPM prolongs sun~val graft less
`effect ivel}' than docs the high dose, but reduces
`
`
`
`West-Ward Exhibit 1055
`Morris 1992
`Page 0029
`
`

`

`thymus weight by 57%. In these animals, thymus
`weight recovers more rapidly 1han in mice treated
`with the higher RPM·dose. The positive correlations
`between RPM dose and prolongation of heart graft
`su1vival and between dose and the duration of
`thymus weight loss suggests that the effects of RPM
`on the thymus contribute"to its immunosuppressive
`efficacy in the mouse.
`ln contrast to the cffecls or treatment wilh high(cid:173)
`or low-dose RPM, treatment of mice with 6 mg/kg
`FK506 produces thymus weight loss of only20%. This
`dose of FK506 prolongs the survival of ear-heart
`grafts longer than low-dose RPM, suggesting that
`involution of the thymus may be more critical to the
`immunusuppressive efficacy of RPM than for the
`efficacy ofFK506.
`Microscopic analysis or thymic tissue from RPM(cid:173)
`treated mice is not complete.Ha Preliminary results
`from hcmatoxylin and eosin staining show that RPM
`treatment disrupts tJie normal thymic architecture.
`The changes in the cortex and medulla are variable,
`but suggest that thymocytes from both areas are
`depleted. lmmunohistochemical staining with mono(cid:173)
`clonal antibodies directed to the pan T cell and
`helper and suppressor/cytotoxic phenotypes has
`shown that RPM causes T-cell depletion; CD4• and
`ens• thymocytcs stain especially weakly in the deep
`cortex. These studies have been performed in collab(cid:173)
`oration with iuvest_igators at Stanford University,
`Stanford, CA and with W.E. Beschorner at Johns
`Hopkins Univl.'rsity, Baltimore, MD. Recently, we
`have found that the cortex is depleted in rats treated
`IV with high doses of RPM. For reasons that arc
`unclear, this effect differs from the thymic medullary
`atrophy caused by treatment with lower doses of
`RPM.63
`The high frequency or cw• /CDS+ double(cid:173)
`positive thymocrtcs in the normal thymus makes
`interpretation o[ the immunohistoehcmic.al staining
`with single antibodies difficult. Therefore, to Learn
`more about the effects of RPM on thymocytes we
`have begun to use two-color Row cytometry. The
`result~ of these studies are complex and cannot yet
`be discussed in detail.,.. In brief, the percentage of
`CD4+ /CDW cells is decreased by RPM treatment
`comparrd with age-matched controls and reaches a
`nadir on day 14. Nthough the percentage of this
`thymocyte subset recovers after cessation of RPM
`treatment, it is still statistically less than normal on
`day 42. RPM also rauses a coincident increase in the
`percent of double- and single-negative (CD4 .. /CD8"
`a nd CD4 .. /CDS-, CD4- /CD+) thymorytrs; thrsr
`
`67
`
`increases probably represent a relative, rather than
`an absolute, increase caused by the primary effect of
`double-positive thymocytc depiction. FK506 treat(cid:173)
`ment causes an effect that is the opposite of RPM:
`the percentage of double-positive cells increases and .
`the percentage of single-positive thymoC)rtcs de(cid:173)
`creases.
`Treatment of mice with RPM also produces com(cid:173)
`plex changes in the response of spleen cells to T- and
`B-cell mitogens in culturellll (Morris RE, Shorthousc
`R: unpublished obse1vations, 1990). Even after cell
`washing in preparation for culture, spleen cells are
`hyporesponsivc to stimulation with STM during and
`after treatment. By day 42, this response returns to
`normal. Thus, the antiproliferative effects of RPM
`are not solely restricted to T cells. In contrast to the
`effects of RPM treatment on the response of spleen
`cells to STM, the response to ConAis not suppressed.
`In fact, spleen cells from RPM-treated mice seem to
`be more responsive lo ConA than spleen cells from
`age-matched control mice. Furthermore, this hyper(cid:173)
`responsiveness increases with time after the cessa(cid:173)
`tion of RPM treatment.
`Recently, we have found that RPM treatmept a lso
`suppresses the increase in PLN weight caused by the
`injection of STM into the hind foot pads of mice
`(Morris RE, Shorthouse R: unpublished observation,
`1991). These results confirm the finding that RPM
`treatment of mice suppresses the response of spleen
`cells in vitro to stimulation by ST.tvI. When succiny(cid:173)
`lated ConA is injected, RPM treatment suppresses
`the PLN response (Morris RE, Shorthouse R: unpub(cid:173)
`lished observations, 1991). Taken together, the data
`from all PLN assays (HvG, GvH and stimulation by
`T- or B-ccll mitogens) show that RPM suppresses
`activation of immune cells stimulated by a variety of
`activation signals in vivo. Thus, RPM seems to cause
`al lca~t two major effects that contribute to its
`im munosuppressive efficacy: (I) reversible depiction
`ofimmature thymocytes; and (2) functional inactiva(cid:173)
`tion, but not substantial depletion, or lymph node
`and spleen cells. The hyperresponsiveness of spleen
`cells to stimulation by ConA in vitro long after RPM
`treatment has ceased may be caused by immature T
`cells emigrating from the rrgrnerating thymus to
`the spleen.
`Since the original comment by Martel ct al,19 the
`effects of RPM treatment on central l)~nphoid tissue
`in large animals have been described by other investi(cid:173)
`gators.6!""72 In our stud)' of monkeys treated for long
`periods \\~th immunosuppressive doses of' RPM (dis(cid:173)
`cussed in section headed Effects of RPMs on Graft
`
`-
`
`
`
`West-Ward Exhibit 1055
`Morris 1992
`Page 0030
`
`

`

`68
`
`Rnndnlf f.1/ii .\lu1ril
`
`and T issue Rcjeclion), lhc rc was some depictio n o r
`cells in lrmph nodes, but the histology of Lhc spleen
`was essentially normal. Because these monkeys were
`juveniles when their RPi\I treatment began, it was
`difficult to ascribe the lack of th~1llllS !issue al
`necropsy entirely to the effects of RPM.
`The inves tigations of the effects of RP:-.·1 on the
`composition or murine thymus a nd spleen and the
`immune function or cells from the spleen and lymph
`node hnve just barely begun to scratch the surface of
`the complex effects ofRP~l on tissues oft he immune
`system. Despite our present naivete, available data
`show that RPM affects centra l lymphoid tissues
`dilfercntly than CsA and FK506. Both CsA"""" and
`11 cause less thymic involution and affect
`FIG06' 11
`"'
`the thymus more selectively than RPi\l because their
`effects are restTicted to depiction of thymocytes in
`the medulla a nd not the cortex. Brief treatment with
`d thcr CsA or F K506 at doses t hat exceed those
`needed for immunosuppression causes reversible
`thymus weight loss. Prolonged treatment with CsA
`renders the thymus incapable of recovery. Both CsA
`and FK506 seem lo mediate their clfccts directly or
`indirectly by damaging medulla.ry e pitheli um. T his
`effect, perhaps in addition to othe rs, may inrcrru pt
`the maturation of singlc-positi,·e CO.·!-• /CDS", CD+·;
`CDS' th)mOC}tC subsets from their double-positive
`
`precursors, cause· a11 increase in the proportiun of
`double-positive cells, and rcducr thr migration of
`crlls from the cortex to the medulla.
`Although the unusual effects of RPi\I on the
`morphology and th)111ocyre subset composition can(cid:173)
`not br ful ly explained until m ore is k nm1~1 about its
`actions, we can speculate on mechanisms that might
`be rt·sponsiblc !'or the effects of'RPf\I (Fig 8). Ir RPM
`u·catmcm docs not interfere with clonal deletion ol
`double-positive cells b)· apoptosis but docs block the
`rescue of double-positive thymOC)1es from cell death
`(rosith·c srlcction), a net loss or double-positive
`th)mocytcs cells will occur. T reatm e nt with FK506
`(or CsA) will have the opposite effect because FK506
`(or CsA) will prevent ncgati\·e selection of potentially
`autorcactive double-positive cells by programmed
`cell death. This tentati,·c hypothesis may explain why
`animals bric Ay trl'atcd with RPM do no t develop the
`syndrom e or S)'l1gcneic GvH (Zhrng B, Morris RE:
`unpublished obser\'ation, 1991) that has been dr(cid:173)
`scribcd in animals treated \dth CsA.'"I."' Three lines
`or C\idence"'·111 support this hypothesis: (I) RPf\!
`dot's not inhibit acth·ation-induced hybridoma apop(cid:173)
`losis and cell dra th in '~tro; (2) CsA inhibits D NA
`fragme ntation in immature thyi110C)'lt'S, and FK506
`inhibits activation-induced apoptosis of hybridoma
`cells; and (3) recc111 Ir it has been suggested that the
`
`co4-cos-
`FK506 ~ ~ - ~~~ +
`~
`~~~
`
`co4+cos+
`~
`~~~
`~~ ~
`
`co4+co1n co4-cos+
`
`~ ~
`~~
`~
`
`•
`
`RPM ~~~ + ~~ • ~ ~ -
`
`~~
`
`~~~~
`~ ~
`
`~
`
`~~
`
`+
`
`CELL DEATH
`
`+
`
`CELL DEATH
`
`~
`~~~
`~~ ~
`
`Figure 8. Possible crrfclS of trt.·atmtnl with FK506 OJ Rl'.\I Ult (he intrathymic differentiation nf thymncyces. Flow
`q •tornctric analyses ol t h)'m<>cytrs from FKjOG·ltTatt>d min· and R.l'.\1-tri•ated mile show decreast<I proportions of
`si11gk-positivc {CIH+COH- /CD+- CDH+) and doublc-positiw cells (CD·I +C:DIH ), respccti\'d)" 1\l though other
`rxpl:111atio11s a re possible (increased cell dC<llh of single-positive l't0 lls or the ir acl'dcraled migrntiun [lmlrl 1111111 .. j into
`the pc1 ipher)'}, the clfcrl of' FK506 I rrallncnt 1s most likdy rnust'cl hy .111 intt·rrupliun in the maturation of
`douhlt•·positive thymot.ytes. Failure.' to bl· rescued from p11si1iH· selc«lion ri·sulting in incrt':.t.,t·d cell death is the
`m1M hkcl) rxplanation fnr lhl' drrrl'.isr in duublc-positin: th) morytes in mice I rea11·d wi1 h RPl\1; an intaruptiun of
`m,llur:uion from dnublc-nrgath·I' crlls ur an accelerated diffrrentiatinn from double-positive n·lls into ~inglt-­
`pnsith-r t hpnnC) tes could also explain the clfects of RP:'ll u catmcnl.
`
`
`
`West-Ward Exhibit 1055
`Morris 1992
`Page 0031
`
`

`

`Rofx1119ri11r
`
`69
`
`CsA/FK.506-resistant CD28signal transduction path(cid:173)
`\\':\} in t hytnot')'tCs is necessal)' for positi\'c selection
`in the thymus"" and RP~l inhibits l}mphocytc activa(cid:173)
`11111
`tion in \~tro via this pathway.' '.,.2
`''
`•
`Other phenomena mighL explain the effects of
`RPM treatment on th)1nocytc subpopulations. J:or
`example, RP~! could be directly toxic to double(cid:173)
`positive cells and cause increased cell death. A less
`likely explanation for I hr low percentage of double(cid:173)
`positi\'C cells is that RPM accelerates the differentia(cid:173)
`tion of' this immature population to cells bearing the
`mur.e mature single-positive phcnot)vc. Although
`the percentage of singlc-positi,·c cells increases dur(cid:173)
`ing RPM treatment, the substantial loss of cells in
`the thymus caused br RPl\'l favors a net cell loss of
`thymocytcs rather than their redistribution br accel(cid:173)
`erated maturation from the cortex to the medulla.
`Because the increase in the percent of double(cid:173)
`ncgativc thymocytes in RPM-treated mice is not
`quantitative!)· im'Crsclr proportional to the decrease
`in dnuble-po~itive cells, it is unlikely that RP:\l causes
`a decrease in the percentage of the double-positi\'e
`population solely br arresting the maturation of
`dnuble-negati\-c thymOC)'tCs to the double-positi\·e
`phenotype.
`Our preliminar)' findings on the distinct effects of
`RP.to.I on murine th)1110C)te populations combined
`with our studies in vivo on the acquisition of specific
`unresponsiveness to ear-heart grafts suggests that
`RP.to.I treatm;:nt ma>' Pl"O\ide the appropriate en\iron(cid:173)
`ment for the induction of tolerance. RPM, acting on
`mature circulating T cells, could prevent immediate
`graft rejection. RPM, acting on thymocytes, could
`enable maLuring and potentially alloreactive th)'mo(cid:173)
`C)'lCS in the recovering thymus to be ncgativelr
`sclrctcci when donor m<\jor histocom patibility com(cid:173)
`plex (l'vlHC) peptides are presented by thymic den(cid:173)
`dritic cells in the context of self-l'vlliC. However, we
`have recently found that RPM treatment of adult
`thymectomized recipients of ear-heart allografts also
`causes prolongation of graft survival (Morris RE,
`Shorthousc R: unpublished obsen."ations, 1991). Al(cid:173)
`though prolongation of graft sun.frat in thrse mice is
`not a.s great as in RP:\1-treared euthymic mice, any
`hypothesis or the mechanism of immunosuppression
`of RPM may have tu be expanded to include clonal
`anerm and acti\·c suppression.
`In the future, the pharniaculugical effects ofRPi\l
`on the th)111llS and other primal)· l~111phoid tissues
`may prodde Yaluable dues to define rhc events
`leading lo self and non-self di